Overtopping hazard on a rubble mound breakwater

A major concern of coastal engineering is not only to access the damage to coastal structures by severe wave overtopping, but also the hazard imposed to users. Local hazard is often associated to the volume of overtopping water per unit of time (called overtopping discharge). Despite two decades of intensive research, it is yet not fully clear to practitioners what is the best method to compute the discharge parameter and its application on the assessment of local hazard. This work provides insight into the overtopping characterization in rubble mound breakwaters, by distinguishing different methods to assess hazardous overtopping. Fieldwork was conducted over a tidal cycle in a breakwater located at Albufeira Harbour (South coast of Portugal) under storm conditions (Hso similar to 3 m; Tp similar to 9 s). Mean overtopping discharges were calculated from field measurements of flow depths and velocities at the breakwater slope, armour and at the impermeable crest. Two different velocities were calculated: overtopping leading-edge velocity and overtopping peak velocity. The two methods provided similar results, with higher velocities occurring during high-tide (between 2 and 10 m/s). Mean overtopping discharges at the beginning of the impermeable crest ranged between 0.2 and 0.8 l/s/m. Under the measured hydrodynamic conditions, the breakwater offers risk to all types of pedestrians. Additionally it is shown that field measurements compare relatively well with empirical prediction methods (for the overall analysed overtopping events), namely the corrected NN_OVERTOPPING2 neural network tool. Besides contributing to the overall database on wave overtopping in coastal structures, the presented results can also be used for calibration and validation of overtopping evaluation methods (empirical formulae, artificial neural networks and numerical and physical models).